Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)


This research has been an experimental investigation of some of the operational aspects of a hypothetical multiproduct, multistage production inventory system operating in a supply uncertainty environment. The main objective of this study was to explore the relative effect of different multilevel buffering strategies on system performance in order to establish some guidelines for choosing among different buffering techniques when buffering the system against different conditions of supply uncertainty. Several performance criteria, including holding cost, inventory cost, total cost, number of shortages, number of stockouts, service level and buffering cost effectiveness, were used to evaluate system performance. The independent variables investigated include: buffering strategy (6 strategies), type of supply uncertainty (4 levels), and degree of supply uncertainty (4 levels). Five replications were generated for each of the 96 cells in the three-factor, full factorial experimental design. The main effect for each factor and the interaction effect for different combinations were considered. Results show that performance of the production system is significantly influenced by the "buffering strategy" factor, although the relative impact of the six buffering strategies is dependent on the performance measure considered. The study also shows that both uncertainty types (quantity and timing) and uncertainty level (high and low) have significant impact on system performance. Moreover, interaction between buffering strategy and either uncertainty type or uncertainty level, were also found to be important in several cases. Overall, this research provides empirical evidence that both supply uncertainty type and level are significant decision variables regarding the selection of an appropriate buffering strategy.